1. Field of the Invention
The present invention relates to an image-shake correcting device having correcting means for correcting image shakes.
2. Description of the Related Art
Modern cameras automatically perform all important operations for photographing such as exposure determination and focusing, so that even unskilled camera operators are unlikely to make mistakes in photographing.
Further, image stabilizing systems have recently been studied, which prevent a photograph from being influenced by shakes of a camera, thereby substantially eliminating factors that induce the photographer""s mistakes in photographing.
Here, an image stabilizing system for cameras will be described.
Unwanted shakes of a camera during photographing are vibrations typically having a frequency of 1 to 10 Hz. A basic concept of obtaining photographs free from image shakes even if the camera shakes at the time of shutter release is that vibrations of the camera resulting from shakes thereof are detected, and a correcting lens is displaced based on the detected vibration value. Therefore, to take photographs having no image shakes or image blurs even with shakes of the camera, first, vibrations of the camera must be accurately detected, and secondly, changes in the optical axis due to shakes of the camera must be corrected.
A basic system for detecting vibrations or shakes of a camera is realized by a shake detecting device installed in the camera, which is comprised of a shake sensor for detecting acceleration, angular acceleration, angular velocity, angular displacement, or the like, and an arithmetic section that arithmetically processes an output from the shake sensor, for camera-shake correction. Based on the detected information, correcting means that decenters the photographic optical axis is driven to suppress image shakes.
FIG. 8 is a schematic perspective view generally showing a camera equipped with a conventional image stabilizing system. This image stabilizing system has a function of performing shake correction for vertical and horizontal shakes of the camera as shown by arrows 42p and 42y, respectively, with respect to an optical axis 41.
In the camera 43, reference numerals 43a, 43b, 43c, and 43d denote a release button, a mode dial (including a main switch), a retractable flash unit, and a finder window, respectively.
FIG. 9 is a perspective view showing the internal construction of the camera in FIG. 8. In FIG. 9, reference numerals 44, 51, and 52 denote a camera main body, correcting means, and a correcting lens, respectively. Reference numeral 53 denotes a support frame that freely drives the correcting lens 52 in directions 58p and 58y, shown in the figure, to execute shake corrections in the directions shown by the arrows 42p and 42y in FIG. 8. The correcting lens 52 will be described later in detail. Reference numerals 45p and 45y denote shake detecting devices such as an angular velocity meter and an angular acceleration meter which detect shakes of the camera in directions 46p and 46y. 
Outputs from the shake detecting devices 45p and 45y are converted into a drive target value for the correcting means via arithmetic devices 47p and 47y, described later. The drive target value is input to coils provided in the correcting means 51 for shake corrections. Reference numerals 54, 56p and 56y, and 510p and 510y denote a base plate, permanent magnets, and the coils, respectively.
FIG. 10 is a block diagram showing the details of the arithmetic devices 47p and 47y in FIG. 9. The arithmetic devices 47p and 47y are constructed similarly to each other, and therefore in FIG. 10, only the arithmetic device 47p is shown and will be described.
The arithmetic device 47p is shown enclosed by one-dot chain lines and comprised of a DC cut filter 48p, a low-pass filter 49p, an analog-to-digital conversion circuit (hereafter simply referred to as the xe2x80x9cA/D conversion circuitxe2x80x9d) 410p, and a drive device 419p, as well as a camera microcomputer 411 which is enclosed by broken lines. The camera microcomputer 411 is comprised of a storage circuit 412p, a differential circuit 413p, a DC cut filter 414p, an integrating circuit 415p, a storage circuit 416p, a differential circuit 417p, and a PWM (Pulse Width Modulation) duty changing circuit 418p. 
In the illustrated example, the shake detecting device 45p is comprised of a laser gyro that detects the angular velocity of shakes of the camera 43. The laser gyro is driven in synchronism with turning-on of the main switch of the camera to start detecting the angular velocity of shakes of the camera 43.
An output signal from the shake detecting device 45p is subjected to cutting-off of DC bias components superimposed on the signal by the DC cut filter 48p composed of an analog circuit. The DC cut filter 48p has such a frequency characteristic that frequencies of 0.1 Hz and less are cut off, and thus does not affect the frequency band of shakes of the camera, which typically ranges from 1 to 10 Hz. However, a problem with the characteristic that frequencies of 0.1 Hz and less are cut off is that about 10 seconds elapse after a shake signal has been input from the shake detecting device 45p and before the DC components are completely cut off. Thus, a smaller time constant is used for the DC cut filter 48p (the characteristic is set such that for example, frequencies of 10 Hz and less are cut off) before, for example, 0.11 seconds elapse after the main switch of the camera is turned on, so that the DC components are cut off in a short time such as 0.1 seconds, and then the time constant is increased (the characteristic is set such that the frequencies of 0.1 Hz and less are cut off), thereby preventing the DC cut filter 48p from degrading a shake angular velocity signal from the shake detecting device 45p. 
An output signal from the DC cut filter 48p is amplified by the low pass filter 49p composed of an analog circuit, at an amplification ratio according to the resolution of the A/D conversion circuit 410p, while high frequency noise components superposed on the shake angular velocity signal are cut off. This cutting-off of high frequency noise components is carried out to prevent the A/D conversion circuit 410p from erroneously sampling the shake angular velocity signal input to the camera microcomputer 411. Further, an output signal from the low pass filter 49p is sampled by the A/D conversion circuit 410p and the resulting digital signal is delivered to the camera microcomputer 411.
As noted above, the DC bias components are cut off by the DC cut filter 48p. However, the subsequent amplification by the low pass filter 49p causes DC bias components to be again superposed on the shake angular velocity signal. Therefore, the DC bias components must be cut off again in the camera microcomputer 411.
Thus, for example, the DC components are cut off by storing, in the storage circuit 412P, a shake angular velocity signal which is sampled 0.2 seconds after turning-on of the main switch of the camera 43, and determining a difference between the previously stored value and the newly stored shake angular velocity signal by means of the differential circuit 413p. This operation can only roughly cut off the DC components because the shake angular velocity signal stored 0.2 seconds after turning-on of the main switch of the camera 43 contains not only DC components but also actual shake components. Therefore, in a subsequent stage, DC components not removed by the differential circuit 413p are completely cut off by the DC cut filter 414p composed of a digital filter. The time constant of the digital filter 414p can be varied as is the case with the analog DC cut filter 48p. Specifically, the time constant is progressively increased over 0.2 seconds after the lapse of 0.2 seconds have elapsed after turning-on of the main switch of the camera 43. More specifically, the filter characteristic of the DC cut filter 414p is preset such that frequencies of 10 Hz and less are cut off when 0.2 seconds have elapsed after turning-on of the main switch, and subsequently the cutoff frequency of the filter is gradually reduced to 5, 1, 0.5, and 0.2 Hz every 50 msec.
However, if during the above-described operation, the photographer half-depresses the release button 43a (turns on a switch S1) to perform a photometric operation or a distance measuring operation, he is likely to immediately carry out photographing. In such a case, it is not desirable to change the time constant over a considerable time. Therefore, in such a case, the change of the time constant is interrupted depending upon the photographing conditions. For example, if the results of the photometric operation indicate that the shutter speed should be {fraction (1/60)}, and the focal distance is 150 mm, then high shake-correcting precision is not required, so that the change of the time constant for the DC cut filter 414p is stopped when the time constant is changed to such a characteristic as to cut off frequencies of 0.5 Hz and less (the amount of change of the time constant is controlled according to the product of the shutter speed and the shooting focal distance). This reduces the time required to change the time constant, thus giving priority to the shutter chance. Of course, it may be so designed that with a higher shutter speed or a shorter focal distance, the change of the time constant of the DC cut filter 414p is stopped when the time constant is changed to such a characteristic as to cut off frequencies of 1 Hz and less, whereas with a lower shutter speed or a longer focal distance, shooting is inhibited until the change of the time constant to the set greatest value is completed.
The integrating circuit 415p starts integrating output signals from the DC cut filter 414p in response to half depression of the camera release button 43a (turning-on of the switch S1), to convert the angular velocity signal into an angular signal. However, so long as the change of the time constant of the DC cut filter 414p has not been completed yet, the integrating circuit 415p does not perform the integration operation until the change of the time constant is completed, as described previously. Although not shown in FIG. 10, the integrated angular signal is amplified at an amplification ratio according to the current focal distance and object distance information and converted so as to drive the correcting means 51 by an appropriate amount according to the angle of shakes. This correction is required because a zoom focusing operation causes a change in the photographic optical system and hence a change in the amount of eccentricity of the optical axis with respect to the amount by which the correcting means 51 is driven.
When the release button 43a is fully depressed (a switch S2 is turned on), the correcting means 51 starts to be driven in accordance with the shake angular signal. However, at this time, care must be taken such that the correction means 51 does not suddenly start a shake correcting operation. The storage circuit 416p and the differential circuit 417 are provided for this purpose. The storage circuit 416p stores the shake angular signal from the integrating circuit 415p in synchronism with the full depression of the release button 43a (turning-on of the switch S2). The differential circuit 417p determines a difference between the signal from the integrating circuit 415p and a signal from the storage circuit 416p. Thus, when the switch S2 is turned on, the two signals input to the differential circuit 417p are equal to each other, and a drive target value signal supplied to the correcting means 51 from the differential circuit 417p is zero. However, the output from the differential circuit 417p subsequently consecutively increases from zero. That is, the storage circuit 416p plays a role in setting the integration signal as the origin when the switch S2 is turned on. This prevents the correcting means 51 from suddenly starting to be driven.
The target value signal from the differential circuit 417p is input to the PWM duty changing circuit 418p. When voltage or current corresponding to the angle of shakes is applied to the coil 510p (see FIG. 9) provided in the correcting means 51, the correcting lens 52 is driven correspondingly to the angle of shakes. PWM driving is preferably used to save power consumed to drive the correcting means 51 and a drive transistor for the coil.
Thus, the PWM duty changing circuit 418p changes the coil drive duty according to the target value. For example, in the case of PWM using a frequency of 20 KHz, the duty is set to xe2x80x9c0xe2x80x9d when the target value from the differential circuit 417p is xe2x80x9c2,048xe2x80x9d and to xe2x80x9c100xe2x80x9d when the target value is xe2x80x9c4,096xe2x80x9d. Then, the range between the duty of xe2x80x9c0xe2x80x9d and the duty of xe2x80x9c100xe2x80x9d is divided at equal intervals so that the duty is determined according to the target value. The duty determination precisely controlled based not only on the target value but also on the current photographing conditions for the camera 43 including temperature, the position of the camera, and the state of the power supply, so as to achieve precise shake corrections.
An output from the PWM duty changing circuit 418p is input to the drive device 419p, which may be a known device such as a PWM driver, and an output from the drive device 419p is applied to the coil 510p (see FIG. 9) provided in the correcting means 51 to carry out shake corrections. The drive device 419p is turned on in synchronism with turning-on of the switch S2 and is turned off once the exposure to the film is completed. Further, even after the exposure has been completed, the integrating circuit 415p continues the integration operation so long as the release button 43a is half-depressed (the switch S1 is on). Then, when the switch S2 is then turned on, the storage circuit 416p again stores a new integration output.
When the half depression of the release button 43a is stopped, the integrating circuit 415p stops integrating outputs from the DC cut filter 414p and is reset. The term xe2x80x9cresetxe2x80x9d, as used herein, refers to an operation of erasing all integrated information.
When the main switch is turned off, the shake detecting device 45p is turned off to complete one image stablization sequence.
Further, when the output signal from the integrating circuit 415p becomes larger than a predetermined value, then it is determined that the camera 43 has performed a panning operation, and the time constant of the DC cut filter 414p is changed. For example, the time constant is changed so that the characteristic that frequencies of 0.2 Hz and less are cut off is changed to one that frequencies of 1 Hz and less are cut off, and is then returned to its original value over a predetermined time period. Specifically, when the output signal exceeds a first threshold, the characteristic of the DC cut filter 414p is set so as to cut off frequencies of 0.5 Hz and less. When the output signal exceeds a second threshold, the characteristic of the DC cut filter 414p is set so as to cut off frequencies of 1 Hz and less. If the output signal exceeds a third threshold, the characteristic of the DC cut filter 414p is set so as to cut off frequencies of 5 Hz and less.
Further, when the output from the integrating circuit 415p becomes very large, the integrating circuit 415p is reset to prevent arithmetic overflow.
In FIG. 10, the DC cut filter 414p starts operating 0.2 seconds after the main switch has been turned on. However, the present invention is not limited to this, but the DC cut filter 414p may be set to start operating when the release button 43a is half-depressed. In this case, the integrating circuit 415p is started to operate when the change of the time constant of the DC cut filter is completed.
Further, as described above, the integrating circuit 415p also starts operating when the release button 43a is half-depressed (the switch S1 is turned on), but may do so when the button 43a is fully depressed (the switch S2 is turned on). In this case, the storage device 416p and the differential circuit 417p are not required.
In FIG. 10, the arithmetic device 47 is provided therein with the DC cut filter 48p and the low pass filter 49p, but it goes without saying that these components may be provided in the shake detecting device 45p. 
FIGS. 11 to 13 show the details of the correcting means 51 in FIG. 9. Specifically, FIG. 11 is a front view of the correcting means 51 in FIG. 9, FIG. 12A is a side view of the correcting means 51 as viewed from the direction of an arrow B in FIG. 11, FIG. 12B is a sectional view taken along line Axe2x80x94A in FIG. 11, and FIG. 13 is a perspective view of the correcting means 51 in FIG. 9.
In FIG. 11, the correcting lens 52 is fixed to the support frame 53. The correcting lens 52 is comprised of two lenses 52a and 52b (FIG. 12B) fixed to the support frame 53, and a lens 52c fixed to the base plate 54, to constitute a group of photographic optical systems.
A yoke 55 made of a ferromagnetic material is mounted on the support frame 53. Mounted on the yoke 55 are the permanent magnets 56p and 56y made of neodymium or the like, which are attracted to a rear side surface of the yoke 55 as viewed in the figures, as indicated by hidden lines. Further, three pins 53a radially extend from the support frame 53 and are fitted in elongated holes 54a formed in side walls 54b axially projected from the base plate 54.
As shown in FIGS. 12A and 13, the pairs of the elongated holes 54a and the pins 53a fitted therein serve to prevent back-lash in the direction of the photographic optical axis of the correcting lens 52 (indicated by the optical axis 57 in FIG. 12A). However, since the elongated holes 54a are elongated in a direction orthogonal to the direction of the optical axis 57, the pairs of the elongated holes 54a and the pins 53a restrict movement of the support frame 53 relative to the base plate 54 in the direction of the optical axis 57, while allowing free movement of the support frame 5a in a plane orthogonal to the optical axis 57, as shown by arrows 58p, 58y, and 58r. However, a tension spring 59 is engaged between each hook 53b on the support frame 53 and a corresponding hook 54c on the base plate 54, as shown in FIG. 11, thereby elastically restricting the movement of the support frame 53 in each of the directions 58p, 58y, and 58r. 
The coils 510p and 510y are mounted on the base plate 54 in opposed relation to the respective permanent magnets 56p and 56y, as partially shown by hidden lines. The yoke 55, the permanent magnet 56p, and the coil 510p are arranged as shown in FIG. 12B, and the permanent magnet 56y and the coil 510y are similarly arranged. When current is caused to flow through the coil 510p, the support frame 53 is driven in the direction of the arrow 58p. When current is caused to flow through the coil 510y, the support frame 53 is driven in the direction of the arrow 58y. 
The amount of driving of the support frame 53 is determined by the balance in each direction between the spring constant of the tension spring 59 and thrust resulting from the coaction between the coils 510p or 510y and the permanent magnets 56p or 56y. That is, the amount of eccentricity of the correcting lens 53 can be controlled by the amount of current flowing through the coils 510p and 510y. 
Recent compact cameras have been significantly miniaturized, and correspondingly both the length and diameter of taking lens barrels in which taking lenses are fitted have been substantially reduced.
Under these circumstances, to install the above-described image stabilizing system into a camera, it is desirable to further reduce the size of the correcting means 51, described above with reference to FIGS. 11 to 13.
However, it can be anticipated that the reduction of the size of the correcting means 51 will necessitate arranging peripheral mechanisms such as the shutter, lens driving device, or lens barrier, which are magnetic members, at locations closer to the correcting means 51. In particular, if the magnetic members are located closer to the permanent magnets provided in the correcting means 51, the correcting lens, which is moved in unison with the permanent magnets, can show low responsiveness, i.e. lacks accuracy in its movement to degrade the image stabilization performance. The best way to prevent this is to contrive an improved layout of the camera, but the pursuit of the size reduction might bring about worse situations which cannot be avoided simply by improvement of the layout.
It is a first object of the present invention to provide an image-shake correcting device which is capable of performing image shake correction with high accuracy using a simple construction and without reducing the degree of freedom of layout and without the need to increase the size of the device, by arranging a second magnetic member that cancels the effects of the magnetic force of a first magnetic member located close to permanent magnets provided in correcting means for correcting image shakes.
It is a second object of the present invention to provide an image-shake correcting device, which employs elastic members that apply elasticity to a correcting lens in a direction in which the correcting lens is driven, and arranges the elastic member at substantially the same location as a support member that supports the correcting lens, and provides support shafts that support the support member with a function of adjusting the elastic force of the elastic members, thereby reducing the space occupied by the elastic members and hence permitting the device to be designed compact in size at low costs without lowering image-shake correcting accuracy, as well as enabling a device or apparatus in which the present device is mounted to be designed compact in size.
To attain the first object, a first aspect of the present invention provides an image-shake correcting device comprising a correcting optical unit having an optical axis, at least one magnet member provided in the correcting optical unit, at least one coil member arranged away from the magnet member in a direction of the optical axis, a first magnetic member arranged away from the magnet member in the direction of the optical axis, and a second magnetic member arranged away from the magnet member in the direction of the optical axis and at a side of the magnetic member remote from the first magnetic member, and wherein energization of the coil member causes the correcting optical unit to be driven in a direction intersecting with the optical axis to correct image shakes.
Preferably, an electromagnetic attractive force exerted between the magnet member and the first magnetic member is substantially equal to an electromagnetic attractive force exerted between the magnet member and the second magnetic member.
Preferably, the first magnetic member comprises a position detecting element having a metal terminal.
More preferably, the position detecting element is a photo interrupter.
Preferably, the second magnetic member comprises an iron-based metal plate.
In a preferred form of the first aspect, the image-shake correcting device comprises a support member that supports the correcting optical unit in a manner permitting same to move in the direction intersecting with the optical axis, and at least one support shaft provided on the support member and projected therefrom in the direction intersecting with the optical axis, and wherein the correcting optical unit has a fitting portion in which the support shaft is fitted, the correcting optical unit being driven in the direction intersecting with the optical axis in response to sliding of the support shaft in the fitting portion.
Preferably, the support shaft is adjustable in position relative to the support member in a direction of the optical axis.
More preferably, the image-shake correcting device comprises an urging member provided on the support shaft, for urging the correcting optical unit in the direction of the optical axis.
According to the first aspect of the present invention, image shakes can be accurately corrected using a simple construction and without reducing the degree of freedom of layout and without the need to increase the size of the device, by arranging the second magnetic member that cancels the effects of the magnetic force of the first magnetic member located close to the permanent magnet provided in the correcting means.
To attain the second object, a second aspect of the present invention provides an image-shake correcting device comprising a correcting optical unit having an optical axis and at least one fitting portion, a support member that supports the correcting optical unit in a manner permitting same to move in a direction intersecting with the optical axis, and a driving unit that drives the correcting optical unit relative to the support member in the direction intersecting with the optical axis, at least one support shaft provided on the support member, the support shaft being fitted in the fitting portion and projected from the support member in the direction intersecting with the optical axis, the support shaft being mounted on the support member in a manner being adjustable in position in a projecting direction thereof relative to the support member, at least one urging member provided on the support shaft, for urging the correcting optical unit in the projecting direction thereof, wherein an urging force of the urging member can be adjusted by adjusting a position of the support shaft, and wherein a driving force of the driving unit drives the correcting optical unit to cause sliding of the correcting optical unit on the support shaft to correct image shakes.
Preferably, the urging member comprises a compression coil spring fitted on the support shaft.
Preferably, the support shaft is threadedly coupled with the support member.
According to the second aspect of the present invention, the support member that supports the correcting lens and the elastic member (urging member) that applies elasticity to the correcting lens in a direction in which the correcting lens is driven are arranged at substantially the same location, and the support shafts that support the support member are each provided with the function of adjusting the elastic force of the elastic member. This construction serves to reduce the space occupied by the elastic member and and hence permit the device to be designed compact in size at low costs without lowering image-shake correcting accuracy, as well as enable a device or apparatus in which the present device is mounted to be designed compact in size.
The above and other objects, features, and advantages of the invention will be more apparent from the following detailed description taken in conjunction with the accompanying drawings.